Large-Eddy Simulation of Turbulent Airflow and Pollutant Dispersion from a Ground-level Point-Source in a Model Urban Area

This study aims to comprehensively investigate pollutant dispersion within a scaled urban model and assess associated risks from emissions. Specifically, we focus on a ground-level point-source in the first row of buildings, continuously releasing a tracer gas for passive scalar transport analysis. The research seeks to understand flow patterns and pollutant dispersion considering the diverse heights and rooftop configurations typical of urban environments. Turbulence significantly influences pollutant dispersion and airflow around structures, prompting large-Eddy simulation (LES) to quantify these effects within the urban model’s regularly spaced buildings. We utilize the dynamic Smagorinsky subgrid-scale (SGS) model to resolve the instantaneous flow field and passive scalar transport. Artificial turbulent structures are generated at the inlet using the synthetic inflow generator method. The validation shows that, the average deviations from the wind tunnel measurements for Wall A at positions \(x_2/H=0\) $x_2/H=0$ and \(x_2/H=3.79\) $x_2/H=3.79$ are approximately 12.09% and 16.52%, respectively. We found that, as free-stream flow encounters the first high-rise buildings in the urban canyon, high streamwise velocity is experienced, followed by the formation of a wake region around obstacles, causing flow separation due to boundary layer detachment from building rooftops. Pollutants released from the ground-level point-source are transported from primary recirculations to secondary ones through turbulent diffusion and advection until evacuated from the urban area. Velocity and concentration contours reveal that in-canyon vortex dynamics and pollutant distribution are highly sensitive to rooftop configurations. The height and shape of buildings not only influence in-canyon vortex structure, but also determine vortex strength. Furthermore, pollutant dispersion characteristics and pollution levels vary across buildings, with distinct regions near high- and low-rise structures showing differing patterns.